Doing away with verb classes

Makes much more since than grinding down and trying to study hard to learn conjugation

There are only 5 rules for Regular 1 conjugations, two more for irregular, and and one for all Regular 2 verbs. This song might be helpful in remembering what the first 7 are:


I think the best way is probably a bit of both. For common verbs, I don’t even think about conjugating them (except when it comes to unusual forms like させららせられる form), I’ve just seen and used them so many times I know what they are without thinking.

However, I still get a little confused sometimes and, in that case, knowing how to conjugate is useful. If you’re talking and you “forget” the て form of a common verb, you can’t just relearn it in an instant, but, if you know how to conjugate it, you can come up with the correct form after a quick 「ええと。。。」. Similarly, when you learn a new verb (and remember, as you progress you are going to learn hundreds and hundreds of them) you know instantly how to make て and ない form, whereas your method, I assume, involves looking them up somewhere.

So basically my - slightly contradictory - advice would be: get familiar with how to conjugate but wherever possible don’t bother actually doing so.

I don’t think it’s any different. If the new verb ends in -iru or -eru, then just learning its dictionary form isn’t enough with either approach. You have to look up the verb class, from which you can work out (though not “instantly” if you’re like me) the て and ない form; I have to look up the て and ない forms, and don’t care about verb class.

If the new verb doesn’t end in -iru or -eru, then it will be very much like some other verb you’ve already learned. For me at least, this is how my brain works best: reasoning by analogy. Maybe I haven’t learned 飛ぶ but it’s easy to see that it works just like 呼ぶ, which I happen to know. Our brains are great pattern machines; given a few examples, they will latch on to the pattern and apply it easily.

The only problem is that the dictionary form isn’t enough to identify the right pattern. You either need to also learn the verb class, or you need to learn a couple other forms. For me, it’s much easier to remember “呼ぶ/呼んで/呼ばない” than it is to remember that it’s a Class 1 (Godan) verb, then dredge up the mnemonic song, remember that む, ぶ, and ぬ go to んで, so it must be 呼んで. It’s just less mental hoops to jump through.

I guess it’s not for everyone, but I’ve been working with it for less than a day, and I already feel like I’ve got a much better grip on these things than I ever did before. I’ll live with it for a few weeks and let you know how it goes!

1 Like

People probably already know what I’ve written below, but it is good practice for me to coalesce my thoughts and write down the patterns for my benefit, if not for anyone else. Please humor/forgive me :slight_smile:.

One pattern I like when learning Japanese is that (other than ーて and -た forms), all Regular 1 verbs basically conjugate according to the 5 vowel sounds in Japanese. Regular 2 are even easier, they only use one vowel (e) in conjugation. So for Reg 1, if you start with the dictionary form, you can get to any of the other conjugations by changing the vowel on the last syllable of the dictionary form. For Reg 2, slice off the -ru from the dictionary form, and add the appropriate conjugation.

Perhaps this chart will help explain what I mean with examples:

  • there are a few Reg 2 verbs that have an “i” in the second-to-last syllable instead of an “e”, e.g.: いる、おりる、みる.

For ーて and ーた, once you learn the 5 rules for -te form for Reg 1 , and learn the rule for Regular 2 (from the dictionary form, change the final -ru to -te) then ーた is just changing the last vowel from “e” to “a”. At this point, you’ve pretty much got the mechanics of verb conjugation down.

(note: irregular verbs する and 来る forms just have to be memorized. And ある is ない in, oddly enough, the nai form.)

From the chart above, or from the -te or -ta forms, all the other conjugations are made, ie:
Reg 1: change dictionary form last vowel from -u to -eru
Reg 2: change dictionary form final -ru to -rareru

Reg 1: drop -nai from ーない conjugation, add -seru
Reg 2: drop -nai from ーない conjugation, add -saseru

Reg 1: drop -nai from ーない conjugation, add -reru
Reg 2: drop -nai from ーない conjugation, add -rareru (like potential)

causative passive:
Using causative form (which is always a Reg 2 form), change the ending -ru to -rareru.

and so on with presumptive, tara conditional, progressive, etc.

For me, the hard part is remembering all the ways to express uncertainty, future events, completed actions, necessity, change in state, indirect questions, direct quotations, suggestions, desires, etc. However, they are all either based on the conjugated forms based on the 5 vowels, -te, -ta, or on the infinitive (polite form verb - masu, i.e. iki from ikimasu), so at least that part is easy.

As if anything about learning Japanese is easy.


させららせられる form

That is not a real form, sir.

られさせたくなかったら on the other hand

[quote=“thornarm, post:25, topic:17349, full:true”] So for Reg 1, if you start with the dictionary form, you can get to any of the other conjugations by changing the vowel on the last syllable of the dictionary form. For Reg 2, slice off the -ru from the dictionary form, and add the appropriate conjugation. [/quote]I have been following a similar process, but starting with the -masu form. For regular 1 (godan) verbs, all you have to do is find the syllable in front of -masu. Look this up in the い column of the hiragana table. Then look across to find the corresponding syllable in the う column of the hiragana table.

Here is a diagram which shows that the syllable before -masu is on the same row as the syllable at the end of the dictionary form.

For the -nai form (of regular 1, godan verbs), replace the う syllable of dictionary form (or い syllable of -masu form) with the corresponding あ syllable in the hiragana table. Note the exception います becomes わない. See following table:

These tables are from the Try! Japanese Language Proficiency Test N5 (Revised Edition) book.

Here is the Wikipedia hiragana table for you to print.


As someone going through Genki I right now, I agree. 2-3 forms are given most chapters.

To the OP: the reason it is て form vs. た (casual past) form is for just the reason I’ve demonstrated - in Japanese it’s a different kana. Yes, it is only a vowel change, however it is a different kana for native learners. Try to get out of relying on romaji sooner rather than later. I know at least one source that teaches with romaji (Steve on the N5 Nihongonomori playlist), and it can be helpful to English-speaking beginners. But for me (as someone who already knew hirigana going into it) just making the connection it was a vowel change was enough - I didn’t need to go through extra steps of turning into romaji and then back into kana again. (That’s what Steve suggested.)

Anyways, I think it’s an interesting and even beneficial idea to learn 2-3 forms at once. At least for me, it’s been working. (And for a few forms, I feel like is less of a new form, but rather an existing form + a word.)

That is a real form, sir. It means you were made to do something by someone else who was made to make you do that thing by a third person who is in turn being controlled by an all-seeing kingpin at the centre of a giant web of coercion. The kingpin’s name is Bob.


How old are you? There’s a decent chance I’ve known hiragana since you were in diapers. :slight_smile: (1990 or so.)

But in this particular case, I think it’s much easier to represent what’s going on in romaji. The shorter you can make a rule, the easier & quicker it is to remember and apply. I’ll probably represent these rules as (for example) “replace -[w]anai with -eba, or -nai with -reba.” Nice and short.

To do the same thing with hiragana is a lot more fiddly; “replace -[something in the あ column]ない with -[same row in the え column, except change わ to just え]ば, or -ない with -れば.” Yuck.

Anybody with a good grasp of hiragana is going to have no trouble applying the rule written in romaji to proper hiragana… it’s just a much more compact way of representing the same thing. This is the one situation where our alphabet is more convenient than a syllabary: when we need to treat the vowel and the consonant independently.

Oh, I thought that was されされさせられぼぶる form


Damn. You’re right. I always get those two mixed up.


Wait, your instruction held off on the う and る (godan and ichidan) distinction?

Maybe I’m lucky, but my high school study back in 06-08 highlighted that immediately, and our approach to verb conjugation closely mirrored that outlined in the original post, using certain conjugations to arrive at the rules for others.

The only ones I give myself pause over at this point are the causative and passive forms.

If (I) do not want to be able to allow it?

させられたくなかったら is also horrifyingly a real form: “If I don’t want to be made to do (something).”

[quote=“iansacks, post:34, topic:17349, full:true”]
Wait, your instruction held off on the う and る (godan and ichidan) distinction?

Maybe I’m lucky, but my high school study back in 06-08 highlighted that immediately[/quote]

I think your experience is typical. Everybody teaches verbs this way. (Except, now, me.)

No it didn’t. I guarantee every conjugation rule was along the lines of “for class 1 verbs, do this; and for class 2 verbs, do that.” This is the problem I’m trying to solve.

I have a big spreadsheet now which uses this new approach to automatically calculate a bunch of conjugations for a bunch of verbs, with no explicit concept of verb classes — only the three core forms (e.g. iku/itte/ikanai). So I’m confident now that it works. Yes, obviously verb classes work too… but they’ve never worked very well for me, because I can never remember which is which and what all the exceptions are.

So for me, at least, I think this new approach (which needs a name, hmm, how about “tri-form verb conjugation”?) is going to work better.

Ah, I see what the goal here is now. For some reason I equated “classes” with “conjugations” for a second. You’re trying to come up with a system that does away with having to assess the ichidan/godan distinction. My bad; that was a failure of comprehension on my part.

Yes, that’s it exactly.

The distinction is still there, implicit in what the plain past (ない) form looks like… but because that’s an actual word I see and use, it’s easier for me to remember than the ichidan/godan distinction. And the way the rules are formulated, it’s very hard to apply them incorrectly (we always try to replace -anai first, falling back to the second option only if there is no -anai to replace).

Does anyone actually sit there and study these huge charts, and then try to use mnemonics and stuff in real life? This is one of those cases where I think you just have to start using the words, speaking to natives, and if you want to focus on conjugations just ask your speaking partner to point out times you screwed it up. That’s a much more natural way of reinforcing the understanding of the rules. The charts are good for reference when you want to be absolutely sure, not something you should be trying to cram into your head.


That’s pretty much what I’ve been doing — I’ve read the conjugation explanations and charts, nodded wisely to myself, and then basically hoped I’d pick it up through usage. For simple stuff that’s worked well enough, but recently my tutor has started giving me some homework more focused on grammar, which has forced me to admit that my grammar skills suck.

But I think you have an important insight here: it’s much easier to remember stuff you use, and to use stuff you remember. To use stuff you don’t remember, but know how to figure out through some complex algorithm, is really hard (and of course requires remembering the algorithm anyway).

That’s why I’m not bothered by having to remember three common verb forms; you’re going to remember those from using them, and use them because you remember them. You’ll remember other common conjugations too, as you use them… but at least this way, when you do need to work through the conjugation rule for some less common use case, you don’t have to start by remembering which @*#! verb class it’s in.

I’m 32, thanks. Why aren’t you typing in hiragana? Too lazy to install an IME? Sorry, didn’t mean to sound condescending at all. Although, apparently you did? In any case, I’ve been working on learning Japanese since '99/2000. So, maybe not as long as you, but still a long time.

I already agreed with you on that, when I mentioned Nihongonomori. To a point, at least.
I don’t bother with “columns”, I just say “it’s a vowel change”. Probably same thing as you, only more internalized. Somewhere between romaji and hiragana charts (and yet also neither)… and to me, much, much simpler.

This sums up exactly how I feel would be the best way to go about it. Honestly, I haven’t been studying Japanese nearly as long as I have Spanish, but being very proficient in Spanish, I can tell you that Japanese verb conjugation seems like a piece of cake compared to Spanish.

But going to back to your point, whenever I had a question or if I was unsure about how to conjugate (in Spanish), I would just ask a native speaker (which happened to be my wife so it was easy :smiley:). I know that not everyone has the luxury of living with a native speaker but I would highly suggest utilizing some of the resources out there that allows you to speak to native speakers and simply ask them when you are unsure. That’s what I plan on doing. You will learn the forms much better and they will stick in your memory but actually USING them. This is one of those areas where actually speaking the language will increase your proficiency in other areas (such as reading/writing) more than actually reading/writing.